Print Join the Discussion View in the ACM Digital Library The mathematical reasoning performed by LLMs is fundamentally different from the rule-based symbolic methods in traditional formal reasoning.

A new method trains artificial intelligence systems to more reliably solve complex problems that require interpreting both text and images. In tests, AI models trained with this method outperformed ...

Study shows dreams really do solve problems. Scientists doubled problem-solving success by playing sounds during sleep to ...

Do you stare at a math word problem and feel completely stuck? You're not alone. These problems mix reading comprehension ...

Chain-of-Thought (CoT) prompting has enhanced the performance of Large Language Models (LLMs) across various reasoning tasks.

Mastery of tool-assisted hunting requires a cognitive threshold previously thought impossible for canines. The 14-month observation period revealed structural limitations in standard predatory ...

Update: Since this article was published, it has emerged that there was a prior solution to the particular Erdos problem solved here by Davenport and Erdős (1936) and Rogers (in Halberstam-Roth (1966) ...

Over the weekend, Neel Somani, who is a software engineer, former quant researcher, and a startup founder, was testing the math skills of OpenAI’s new model when he made an unexpected discovery. After ...

Abstract: The work deals with the NP-hard problem of scheduling tasks on two machines with the sum-cost criterion. In this problem, each task must be performed sequentially on the first and then on ...

Abstract: Though quite challenging, training a deep neural network for automatically solving Math Word Problems (MWPs) has increasingly attracted attention due to its significance in investigating how ...